LaurentiuGabriel/unreal-game-assets-creation-skill

143

stars

2

commits

Python

primary language

Jul 24, 2026

updated

README

text-to-3d-asset

A Claude Code skill that turns a text prompt or a photo into a game-ready, textured 3D asset by orchestrating local AI services in sequence — then (optionally) imports it straight into an Unreal Engine project.

prompt ──▶ Fooocus (SDXL)  ──▶  image.png
image  ──▶ Hunyuan3D-2      ──▶  textured .glb
.glb   ──▶ Blender          ──▶  .fbx
.fbx   ──▶ Unreal (MCP)     ──▶  StaticMesh / BP racer

It was built and validated end-to-end on a single Windows workstation (RTX 4070 Laptop, 8 GB VRAM / 32 GB RAM), generating engines for an Unreal train‑racer game.

Example

A real run of the pipeline — the prompt below became an SDXL image (Fooocus), then a textured .glb (Hunyuan3D‑2), shown here as a turntable of the generated mesh:

text-to-3d-asset demo: prompt → SDXL image → textured 3D turntable

Prompt: "a vintage green steam locomotive, single centered object, plain white background, studio product render, full side view" — every frame after the title card is an actual output of this skill (image from Fooocus, 3D turntable rendered from the Hunyuan3D‑2 .glb).

What it does

StageToolOutput
1. Text → image (optional)Fooocus SDXL, driven headlessly via the Fooocus-API REST wrapper.png
2. Image → 3DHunyuan3D‑2 (image‑to‑textured‑mesh)textured .glb
3. ConvertBlender headless (glTF → FBX, embedded textures).fbx
4. Import (optional)Unreal Engine via the unreal-mcp serverStaticMesh + Blueprint

If you already have a photo, skip Stage 1 and start at Stage 2.

Contents

SKILL.md                 # the skill instructions Claude Code loads
scripts/fooocus_gen.py   # text → image via the Fooocus-API REST endpoint
scripts/hunyuan_gen.py   # image → textured GLB via the Hunyuan3D-2 gradio API
scripts/glb_to_fbx.py    # GLB → FBX (embedded textures) via headless Blender

Requirements

  • Windows with an NVIDIA GPU (8 GB VRAM works; more is better) and recent drivers.
  • Hunyuan3D‑2 running locally (this skill was set up with the WinPortable build); its texture extensions (custom_rasterizer, differentiable_renderer) compiled against a CUDA toolkit matching its PyTorch build.
  • Fooocus + the Fooocus-API wrapper (only needed for Stage 1 text‑to‑image).
  • Blender 4.x (only for Stage 3 FBX conversion).
  • An Unreal project with the unreal-mcp server (only for Stage 4 import).

Install as a Claude Code skill

Copy this folder into a .claude/skills/ directory so Claude Code discovers it:

# project-scoped (active in one project)
cp -r text-to-3d-asset  <your-project>/.claude/skills/text-to-3d-asset

# or user-scoped (active everywhere)
cp -r text-to-3d-asset  ~/.claude/skills/text-to-3d-asset

Then just ask, e.g. "make me a 3D asset of a red double‑decker bus" and the skill takes over.

⚠️ Important: one service at a time (8 GB GPUs)

Fooocus (SDXL) and Hunyuan3D each want most of an 8 GB card. Keeping both loaded exhausts VRAM + RAM and swaps to disk — a generation that takes ~1 min balloons to ~14 min. The skill therefore stops the other server before generating: stop Hunyuan (:8080) before Stage 1, stop Fooocus‑API (:8888) before Stage 2. Relaunching is cheap by comparison.

Note on paths

SKILL.md and the scripts contain absolute paths specific to the machine they were built on (C:\AI\..., C:\projects\...). Adjust them to your own install locations before use.

Credits

Built with Claude Code. Wraps the excellent open‑source projects linked above — all credit to their authors.

Contributors

LaurentiuGabriel/unreal-game-assets-creation-skill

143

stars

2

commits

Python

primary language

Jul 24, 2026

updated

README

text-to-3d-asset

A Claude Code skill that turns a text prompt or a photo into a game-ready, textured 3D asset by orchestrating local AI services in sequence — then (optionally) imports it straight into an Unreal Engine project.

prompt ──▶ Fooocus (SDXL)  ──▶  image.png
image  ──▶ Hunyuan3D-2      ──▶  textured .glb
.glb   ──▶ Blender          ──▶  .fbx
.fbx   ──▶ Unreal (MCP)     ──▶  StaticMesh / BP racer

It was built and validated end-to-end on a single Windows workstation (RTX 4070 Laptop, 8 GB VRAM / 32 GB RAM), generating engines for an Unreal train‑racer game.

Example

A real run of the pipeline — the prompt below became an SDXL image (Fooocus), then a textured .glb (Hunyuan3D‑2), shown here as a turntable of the generated mesh:

text-to-3d-asset demo: prompt → SDXL image → textured 3D turntable

Prompt: "a vintage green steam locomotive, single centered object, plain white background, studio product render, full side view" — every frame after the title card is an actual output of this skill (image from Fooocus, 3D turntable rendered from the Hunyuan3D‑2 .glb).

What it does

StageToolOutput
1. Text → image (optional)Fooocus SDXL, driven headlessly via the Fooocus-API REST wrapper.png
2. Image → 3DHunyuan3D‑2 (image‑to‑textured‑mesh)textured .glb
3. ConvertBlender headless (glTF → FBX, embedded textures).fbx
4. Import (optional)Unreal Engine via the unreal-mcp serverStaticMesh + Blueprint

If you already have a photo, skip Stage 1 and start at Stage 2.

Contents

SKILL.md                 # the skill instructions Claude Code loads
scripts/fooocus_gen.py   # text → image via the Fooocus-API REST endpoint
scripts/hunyuan_gen.py   # image → textured GLB via the Hunyuan3D-2 gradio API
scripts/glb_to_fbx.py    # GLB → FBX (embedded textures) via headless Blender

Requirements

  • Windows with an NVIDIA GPU (8 GB VRAM works; more is better) and recent drivers.
  • Hunyuan3D‑2 running locally (this skill was set up with the WinPortable build); its texture extensions (custom_rasterizer, differentiable_renderer) compiled against a CUDA toolkit matching its PyTorch build.
  • Fooocus + the Fooocus-API wrapper (only needed for Stage 1 text‑to‑image).
  • Blender 4.x (only for Stage 3 FBX conversion).
  • An Unreal project with the unreal-mcp server (only for Stage 4 import).

Install as a Claude Code skill

Copy this folder into a .claude/skills/ directory so Claude Code discovers it:

# project-scoped (active in one project)
cp -r text-to-3d-asset  <your-project>/.claude/skills/text-to-3d-asset

# or user-scoped (active everywhere)
cp -r text-to-3d-asset  ~/.claude/skills/text-to-3d-asset

Then just ask, e.g. "make me a 3D asset of a red double‑decker bus" and the skill takes over.

⚠️ Important: one service at a time (8 GB GPUs)

Fooocus (SDXL) and Hunyuan3D each want most of an 8 GB card. Keeping both loaded exhausts VRAM + RAM and swaps to disk — a generation that takes ~1 min balloons to ~14 min. The skill therefore stops the other server before generating: stop Hunyuan (:8080) before Stage 1, stop Fooocus‑API (:8888) before Stage 2. Relaunching is cheap by comparison.

Note on paths

SKILL.md and the scripts contain absolute paths specific to the machine they were built on (C:\AI\..., C:\projects\...). Adjust them to your own install locations before use.

Credits

Built with Claude Code. Wraps the excellent open‑source projects linked above — all credit to their authors.

See what people are saying

Contributors

Languages

Python

100.0%